The following background material is not admitted to be prior art to the pending claims, but is provided only to aid the understanding of the reader.
Antibiotics have been effective tools in the treatment of infectious diseases during the last half century. From the development of antibiotic therapy to the late 1980s there was almost complete control over bacterial infections in developed countries. The emergence of resistant bacteria, especially during the late 1980s and early 1990s, is changing this situation. The increase in antibiotic resistant strains has been particularly common in major hospitals and care centers. The consequences of the increase in resistant strains include higher morbidity and mortality, longer patient hospitalization, and an increase in treatment costs. (B. Murray, 1994, New Engl. J. Med. 330: 1229-1230.)
The constant use of antibiotics in the hospital environment has selected bacterial populations that are resistant to many antibiotics. These populations include opportunistic pathogens that may not be strongly virulent but that are intrinsically resistant to a number of antibiotics. Such bacteria often infect debilitated or immunocompromised patients. The emerging resistant populations also include strains of bacterial species that are well known pathogens, which previously were susceptible to antibiotics. The newly acquired resistance is generally due to DNA mutations, or to resistance plasmids (R plasmids) or resistance-conferring transposons transferred from another organism. Infections by either type of bacterial population, naturally resistant opportunistic pathogens or antibiotic-resistant pathogenic bacteria, are difficult to treat with current antibiotics. New antibiotic molecules which can override the mechanisms of resistance are needed.
Bacteria have developed several different mechanisms to overcome the action of antibiotics. These mechanisms of resistance can be specific for a molecule or a family of antibiotics, or can be non-specific and be involved in resistance to unrelated antibiotics. Several mechanisms of resistance can exist in a single bacterial strain, and those mechanisms may act independently or they may act synergistically to overcome the action of an antibiotic or a combination of antibiotics. Specific mechanisms include degradation of the drug, inactivation of the drug by enzymatic modification, and alteration of the drug target (B. G. Spratt, Science 264:388 (1994)). There are, however, more general mechanisms of drug resistance, in which access of the antibiotic to the target is prevented or reduced by decreasing the transport of the antibiotic into the cell or by increasing the efflux of the drug from the cell to the outside medium. Both mechanisms can lower the concentration of drug at the target site and allow bacterial survival in the presence of one or more antibiotics which would otherwise inhibit or kill the bacterial cells. Some bacteria utilize both mechanisms, combining a low permeability of the cell wall (including membranes) with an active efflux of antibiotics. (H. Nikaido, Science 264:382-388 (1994)).
In some cases, antibiotic resistance due to low permeability is related to the structure of the bacterial membranes. In general, bacteria can be divided into two major groups based on the structure of the membranes surrounding the cytoplasm. Gram-positive (G+) bacteria have one membrane, a cytoplasmic membrane. In contrast, Gram-negative (G-) bacteria have two membranes, a cytoplasmic membrane and an outer membrane. These bacterial membranes are lipid bilayers which contain proteins and may be associated with other molecules. The permeability of bacterial membranes affects susceptibility/resistance to antibiotics because, while there are a few molecular targets of antibiotics, e.g., penicillin-binding proteins, that are accessible from the outer leaflet of the cytoplasmic membranes, the principal targets for antibiotics are in the cytoplasm or in the inner leaflet of the cytoplasmic membrane. Therefore for an antibiotic which has a target in the cytoplasmic membrane, in Gram-negative bacteria that antibiotic will first need to cross the outer membrane. For a target in the cytoplasm, an antibiotic will need to cross the cytoplasmic membrane in Gram-positive bacteria, and both the outer and cytoplasmic membranes in Gram-negative bacteria. For both membranes, an antibiotic may diffuse through the membrane, or may cross using a membrane transport system.
For Gram-negative bacteria, the lipid composition of the outer membrane constitutes a significant permeability barrier. The outer layer of this outer membrane contains a lipid, lipopolysaccharide (LPS), which is only found in the outer membrane of Gram-negative bacteria. The lipid layer of the outer membrane is highly organized in a quasi-crystalline fashion and has a very low fluidity. Because of the low fluidity of the lipid layer of the outer membrane, even lipophilic antibiotics will not diffuse rapidly through the lipid layer. This has been shown experimentally, hydrophobic probe molecules have been shown to partition poorly into the hydrophobic portion of LPS and to permeate across the outer membrane bilayer at about one-fiftieth to one-hundredth the rate through the usual phospholipid bilayers (like the cytoplasmic membrane bilayer).
Some antibiotics may permeate through water-filled porin channels or through specific transport systems. Many of the porin channels, however, provide only narrow diameter channels which do not allow efficient diffusion of the larger antibiotic molecules. In addition, many porin channels are highly hydrophilic environments, and so do not efficiently allow the passage of hydrophobic molecules. Thus, the outer membrane acts as a molecular sieve for small molecules. This explains, in part, why Gram-negative bacteria are generally less susceptible to antibiotics than Gram-positive bacteria, and why Gram-negative bacteria are generally more resistant to large antibiotics, such as glycopeptides, that cannot cross the outer membrane.
The cytoplasmic membrane also provides a diffusion barrier for some antibiotics. However, since the fluidity of the lipid layer of the cytoplasmic membrane is higher than that of the outer membrane of Gram-negative bacteria, drugs that show some lipophilicity will be able to permeate through the lipid layer. Other drugs, such as phosphonomycin or D-cycloserine that have very low solubility in a lipophilic environment will cross the cytoplasmic membrane by using a transport system. In this case, though, if the transport system is not synthesized, the bacteria will become resistant to the drug (Peitz et al., 1967, Biochem. J. 6: 2561).
Decreasing the permeability of the outer membrane, by reducing either the number of porins or by reducing the number of a certain porin species, can decrease the susceptibility of a strain to a wide range of antibiotics due to the decreased rate of entry of the antibiotics into the cells. However, for most antibiotics, the half-equilibration times are sufficiently short that the antibiotic could exert its effect unless another mechanism is present. Efflux pumps are an example of such other mechanism. Once in the cytoplasm or periplasm a drug can be transported back to the outer medium. This transport is mediated by efflux pumps, which are constituted of proteins. Different pumps can efflux specifically a drug or group of drugs, such as the NorA system that transports quinolones, or Tet A that transports tetracyclines, or they can efflux a large variety of molecules, such as certain efflux pumps of Pseudomonas aeruginosa. In general, efflux pumps have a cytoplasmic component and energy is required to transport molecules out of the cell. Some efflux pumps have a second cytoplasmic membrane protein that extends into the periplasm. At least some efflux pumps of P. aeruginosa have a third protein located in the outer membrane.
Efflux pumps are involved in antibiotic resistance since, in some cases, they can remove a significant fraction of the antibiotic molecules which manage to enter the cells, thereby maintaining a very low intracellular antibiotic concentration. To illustrate, P. aeruginosa laboratory-derived mutant strain 799/61, which does not produce any measurable amounts of efflux pump is 8 to 10 fold more susceptible to tetracycline and ciprofloxacin than the parent strain P. aeruginosa 799, which synthesizes efflux pumps. Also, null mutants of mexA, the cytoplasmic component of a P. aeruginosa efflux pump, are more susceptible to antibiotics than the wild type.
The physiological role of efflux pumps has not been clearly defined yet. They are involved in drug resistance but they also are involved in the normal physiology of the bacterial cell. The efflux pump coded in the mexA operon of P. aeruginosa has been shown to be regulated by the iron content of the medium, and it is co-regulated with the synthesis of the receptors of siderophores. Siderophores are molecules that are needed for bacterial growth under iron starvation conditions, such as during infection of an animal. They are synthesized in the cytoplasm and exported when the bacterial cell needs iron. Siderophores scavenge iron within the infected animal and return the iron to the microbe to be used for essential microbial processes. Since there is essentially no free iron in the bodies of animals, including the human body, the production of siderophores by infecting bacteria is an important virulence factor for the progress of the infection.
Even organisms normally surrounded by a cell envelope of relatively high permeability can develop resistance by decreasing the permeability of the envelope. When an agent mainly diffuses across the barrier through a specific channel, mutational loss of the channel can be an efficient mechanism for resistance. A "nonclassical" .beta.-lactam compound, imipenem, shows an exceptional activity against P. aeruginosa, mainly because this agent diffuses though a specific channel, OprD, whose physiological function appears to be that of the transport of basic amino acids. However, P. aeruginosa could become resistant to imipenem by simply losing the oprD channel, and currently a large fraction of P. aeruginosa strains isolated from the hospital environment are resistant as a result of this modification. In a similar manner, .beta.-lactam compounds designed to mimic iron-chelating compounds (siderophores) during their transport through the outer membranes are known to select mutants that are defective in the specific transport of these siderophores.
In summary, the above discussion indicates that cellular factors affecting transport (both active and passive transport) of antibiotics into bacterial cells are important components of antibiotic resistance for many bacterial species.